📬 You are reading an Essential Brief executive article. Subscribe for daily 3-minute updates →
Infrastructure & Compute (INFRA)

Perplexity open sources Lily inference engine for Qwen

By Essential Brief Intelligence2026-09-042 min read

⚡ Executive Digest (3-Minute Breakdown)

Perplexity has open sourced Lily, a Rust and Metal-based local inference engine for the Qwen3.6-35B-A3B model on Apple silicon, exposing an OpenAI-compatible chat-completions API for on-device text generation. Unlike general-purpose stacks such as MLX and MLX-LM, Lily targets a single model and hardware family, tightly integrating model structure, execution plans, and custom Metal kernels to extract performance from Apple’s unified memory architecture. Benchmarks on an M5 Max show Lily delivering about 1.23x faster prefill and 1.35x faster decode than MLX-LM, supporting deployable local inference via a lightweight HTTP server and potentially guiding future specialized runtimes.

This update represents a notable development in the Ai sector. Organizations and founders tracking this space should evaluate potential strategic and technical implications on their operations.

⚡ Daily Executive Briefing

Get Daily 3-Minute Executive Digests

No fluff, no clickbait. Concise intelligence delivered to your inbox every morning.

🔒 100% Free. One-click unsubscribe anytime. Zero spam.